(warmwell note. There were many mistakes in the .pdf documents quoted below - probably as a result of OCR scanning rather than poor typing. We have attempted to correct them - but some remain where the sense was not clear. The emphasis is ours.)

Extract "Fortunately, we were able to take the devices and test system into the field in Uruguay in November 2001, where they performed splendidly on farm in a remote area...."

The statement made by Roger Breeze of USDA to the Royal Society Enquiry

link: http://www.royalsoc.ac.uk/inquiry/388.pdf

Roger Breeze [RBREEZE@ars.usda.gov]
22 April 2002 2354
Foot and Mouth Enquiry: For Dr. Geoffrey Findlay

Dr. Fred Brown has drawn my attention to statements provided to the Royal Society's Committee of Enquiry after a meeting with Dr. Chris Bostock, Dr. Alex Donaldson, Dr. Soren Alexandersen and Dr. P. Mellor.
Under "IDiagnostics Paragraph 8" is the statement that "The Laboratory had also tested the Cepheid Smart Cycler. Originally, Dr. Roger Breeze, USDA, had been invited to Pirbright to compare the performance of the Smart Cycler with a lab-based RT-PCR system, but this invitation had not been taken up. Subsequently, in the summer, an employee of the company associated with the device came over to set up the equipment and reagents. Using those reagents, the test had very poor sensitivity, but Pirbright had investigated other reagents and the use of those developed for its own RT-PCR test had resulted in comparable sensitivity to the latter. "

The Agricultural Research Service (ARS) of USDA has been developing real time PCR reagents to detect the List A diseases of the OIE. By February 2001 we had developed and tested in the laboratory a remarkable assay that detected all 7 serotypes of FMD virus and differentiated FMD virus from other vesicular diseases of livestock and from other RNA viruses. This assay was performed on either the Cepheid Smart Cycler or the Idaho Technology RAPID machine: both these machines can be monitored from a distance over the internet.

This RT-PCR test is more sensitive than cell culture. During February and March 2001, we found that this test detected the presence of FMD virus in experimentally infected cattle, sheep and swine before clinical signs of disease were detectable: in other words, this was a preclinical test. The test takes about 60 minutes and sample preparation can be minimal - if vesicular fluid is sampled, for example.

In early March 2001, I telephoned Dr. Donaldson to inform him of the USDA RT-PCR test and offered to make the equipment and reagents available. Now the design and purpose of the two instruments we have chosen and the format of the reagents are specifically suited to performance of diagnostic tests on the farm or close to the site of the problem. There are cheaper instruments that will do the same job at a fixed site in the laboratory but these are not robust enough to use on the farm. Given the extensive validation studies in vitro and in vivo that had already taken place at Plum Island, our expectations were that after a short familiarization period (1 to 2 days) for UK colleagues at Pirbright we would be able to take the devices and tests into the field during the 2 0 0 1 FMD outbreak in cooperation with UK authorities from Pirbright or MAFF.

Such studies would have provided the US and UK with valuable data under field conditions: we would have used these data to support USDA licensure and OIE test approval.

Unfortunately, yet understandably, Pirbright staff were too busy coping with the demands of epidemic control to explore new technology during the spring and summer of 2001 . As a result, my offer to provide the latest diagnostic technology was not taken up.

Fortunately, we were able to take the devices and test system into the field in Uruguay in November 2001, where they performed splendidly on farm in a remote area.

ARS is conducting this FMD research in partnership with Tetracore Inc., a reagents company that has a license to commercialize the technology.

Cepheid and Idaho Technology have no involvement in FMD diagnostic research design or performance - they are solely instrument suppliers.

I have been puzzled by the Pirbright statements that Cepheid provided FMD reagents because Cepheid has no involvement in PCR primer design for Tetracore/ARS and the details of the ARS primers have not been disclosed to Cepheid. I do not know the details of the reagents Cepheid provided to Pirbright but I can assure you that these were not the reagents proprietary to ARS/Tetracore that are used in the US test.

The details of the ARS FMD test will be published June 1, 2002 in the Journal of the American Veterinary Medical Association: I am trying to have a galley proof sent to your enquiry under publication embargo until June 1 so that your committee has the scientific data in hand as soon as possible.

Finally, I should clarify some misleading statements about the ARS FMD test. Test reagents are pre-packaged such that contamination is not a problem. Sample preparation is minimized. We envision these devices being taken onto the farm by federal diagnosticians for the index case and then kept close at hand beyond the farm quarantine boundary thereafter.

I must re-emphasize that the ARS test is intended to be used as an on-farm test (although it can be a central laboratory test) because time and timeliness are the critical issues in stamping out an epidemic.

Much reference continues to be made to the relative sensitivities of cell culture and RT-PCR and to the need for spurious comparisons with lab-based PCR instruments

In our assay, RT-PCR is more sensitive than cell culture. When it is possible to do cell culture on the farm - or use fragile lab-based PCR devices in vehicles - we can have a direct head to head challenge of technologies, but until then, the ARS RT-PCR for FMD is the new standard for on-site diagnosis of FMD. We are pursuing the appropriate licensing of this test by US regulatory authorities and approval by OIE.

extract of letter from Roger Breeze, sent as further evidence to Royal Society Enquiry
This letter was sent also to the Lessons Learned Inquiry and given by hand to Lord Whitty

link:http://www.royalsoc.ac.uk/inquiry/387.pdf "....Our real time PCR assay for foot and mouth disease (W) has been validated in the laboratory: it has proven to be a pre-clinical test for infection in cattle, swine and sheep, it detects all 7 serotypes of FMD virus and differentiates this infection from other viral diseases that cause similar clinical signs.

The test is more sensitive than viral culture and will detect as few as 10 virus particles.

The research paper with these results will appear soon in the Journal of the American Veterinary Medical Association.

We have now begun to use this PCR test for field research and recently conducted a study in Uruguay, looking for carrier animals in a herd that had recovered from a FMD infection. This research is in progress and will be published in due course. It is not a validation of the test so much as it is a demonstration that FMD detection can take place in remote places in a country without a national FMD diagnostic capacity.

Such a study could not have been conducted on site using cell culture. I will be sure to send a copy of the paper to Tony Garland.

Some eight months after we had disclosed the existence of our x;h/lD test to Dr. Donaldson, we read in the Veterinary Record that the Pirbright Laboratory had subsequently established a relationship with Cepheid and conducted some experiments with FpvfB reagents supplied by that company (data published by Alex Donaldson and others in the Veterinary Record, 2001).

I have no idea what those reagents were because the paper does not describe them. But I can be sure that these reagents were not those developed by USDA-ARS and Tetracore because Cepheid does not have this proprietary information.

I hope there has been no confusion in Britain between the Cepheid mystery test and the real time PCR test developed at Plum Island.

I did consider writing to the Veterinary Record at the time to clarify this but decided this was not worthwhile since Donaldson's letter was largely anecdotal and Cepheid clearly had no capability in FMD detection, so it was hard to believe anyone would take his comments seriously: it seemed better to wait for our paper to be published with the supporting data. Correspondence columns are not the ideal place to set out new principles for control of the world's most dangerous animal diseases. When the.Plum Island paper is published the current state of the art in FlMD virus detection will be clear.

Periodically, there are major shifts in technology and history shows that many people have problems adjusting to these - the Luddites are the classical example.

The last FMD outbreak in the U.S. was in 1929 and at that time diagnosis was on-farm and somewhat leisurely. Veterinarians travelled to the farm by train and car and communicated by telegraph with Was.hington DC. There were 1x0 Jab tests available and diagnosis was achieved the farm by inoculating a cow, pig and horse with materials h m suspected cases. The choice was between FAID and vesicular stomatitis infection and differential diagnosis depend upon which combination of animals got sick. When FMD hit Mexico in 1946, there was no laboratory in the Americas where a diagnosis could be made: samples from infected animals were sent by sea to Pirbright. By the time the results were telegraphed back weeks later, the infection was all over Mexico and a 6 year struggle had begun.

When the Plum Island laboratory was opened in 1956, it was the first FMD diagnostic capability in this hemisphere and it seemed miraculous to American farmers that a vet could travel by train and car to a farm, collect samples from a cow suspected of having FMD, take those samples to an airport, and put them on a plane to New York, where someone would carry them by road and sea to a lab from which an answer could be obtained in a few days and communicated by telephone.

I'm sure there must have been some old timers who thought that clinical diagnosis on farm was adequate and new technology was unnecessary, just as there are some now who think new ideas will not replace the old.

We in are the only scientists in the U.S. who can do research with FMD virus and we take this responsibility very seriously. For several years we have been studying major animal disease epidemics overseas to see what lessons we can learn to protect the US. Notable examples are FMD in Taiwan, Japan and South Korea, classical swine fever in the Netherlands and Rift valley fever in Saudi Arabia/Yemen. It is very clear that whatever the value of existing control strategies, including diagnosis by ccU culture in specialised laboratories, these are not effective in preventing catastrophic disease losses, even in small countries.

We do not accept that a national catastrophe every generation is the necessary price of having livestock populations susceptible to FMD or classical swine fever. We are not prepared to continue with traditional approaches that do not work. We do believe that we can develop new technologies to prevent and control FMD in the U.S. First, some facts: the distance ji-from New York to San Francisco is the same as that from London to Baghdad. U.S. livestock susceptible to FMD number about 200 million. U.S. agriculture is remarkably mobile and decades of freedom from FMD and other serious diseases have accustomed us to being able to transport animals and products nationwide in hours. The U.S. is unique in what it stands to lose from a FMD epidemic and in the scope of the response that will be necessary to control an outbreak.

Also, our agricultural industries are arranged quite differently from other countries.

In California, 2000 cow dairies are not unusual and these are clustered such that there are hundreds of thousands of dairy cows within a small radius. Farmers milk their animals continuously, 24 hours a day: a succession of milk tankers is filled and sent to the processing plant, there is no milk storage capacity on farm m . If there were a suspected FMD case and a 3 kilometer quarantine standstill area was imposed, there would be an immediate and major crisis over what to do with dens of thousands of gallons of milk - no place to store it on farm, no way to get it to the processing plant, no way to spill it on the ground or pour it down the drains because of environmental pollution. It took only two hours to take samples to Plum Island to get a diagnosis, this would still be two days of chaos on farms in the quarantine zone. We believe answers need to be found in minutes or hours.

.... time is the critical element in FMD control. We do not pretend that we can always hope to prevent a FMD outbreak in the U.S., but what we want to be able to do is to ensure that the outbreak follows the curve shown in line C, rather than line A. Integral to following line C is that infected herds are quickly identified and slaughtered to prevent the infection spreading to others. We could not hope to do this in the U.S. by taking samples to Plum Island - it would simply take too long and cause too many other problems, as in California.

One would like to believe that in an epidemic all the infected herds are detected by cell culture, perhaps via two passages in the most sensitive tissue culture systems, as Tony Garland points out.

The reality is very different - there is usually no laboratory data at all when the decision is made to slaughter and this decision is dictated more by geography than virus detection by any method. Many people have asked me recently if there are any cattle or sheep naturally resistant to FMD - but how would we know this when any resistant animals& would be killed along with all the rest because they were in the wrong place at the wrong time - the wrong place being within 3 kilometers of an outbreak

I do not believe U.S. farmers would accept that their life's work - bloodlines developed over generations in many cases - should be wiped out without clear evidence of infection.

Nor do thy have to.

I must now digress a little about the PCR test. PCR is well established in the laboratory. The portable devices we are using offer all the power of the lab but at the site of of the problem.

There is no issue of cross contamination or distinguishing live from dead FMD virus: there is no FMD virus, live or dead on farms in the U.S.

If the PCR test finds virus on a farm it can only be that there is a serious animal health problem in progress. Much has also been made of the so-called comparison between results from the portable real time PCR device and those from culture. There is no comparison. When it is possible to do cell culture in a portable format on the farm we can arrange a head to head test, but until then, the portable real time PCR test is the only way to identify FMD virus definitively on site. Our test system is NOT intended for laboratory use - there are cheaper devices that do much the same thing. It is designed for rapid results at the site of the problem.

This raises the key question: what would one do differently to control FMD if detection could be provided in minutes on site rather than after days of waiting?

If there is nothing one would do differently, then real time PCR has no value. We believe very strongly that there are many new opportunities after immediate detection and these can be achieved through a command and control program like "LEADERS" that will allow all available resources to be deployed in a science-driven manner in the first hours and days after infection is detected. I won't get into this subject here, it is too lengthy, but it is central to our strategy of nipping infection in the bud and following line C above. It does occur to me that the current debate in Britain has the wrong focus - on whether specific technologies ars valid rather than on what disease outcomes are desired - and it is from desired outcomes that new technologies will flow. For example, we desire that an epidemic follow Iine C above - which requires rapid detection and slaughter of infected herds and flock. With this as a desired outcome, local rapid detection capacity is essential so that in the first few hours after finding FMD on a farm we could sample every herd in the vicinity for FMD - not just those in a set radius - and deal with those that were infected within 24 hours. This might be done without going on the farm if milk heading to the processing plant were sampled. There is no right or wrong answer about the value of rapid PCR versus cell culture - it depends on what type of epidemic you are prepared to live with. If line A is adequate, new technologies are not needed. We are currently proceeding with the research that will lead to regulatory approval of this new generation of tests, first by USDA and then by OIE. This will not be a quick process. Current U.S. policies for control of FMD arc largely the same as those in U.K. Our job in research is to provide new tools that OIE regulatory agencies can add to their arsenal if they choose. Lack of current validation is not the issue. As I said above, the first thing is to decide what kind of outbreak you can live with - and if you want line C, new technologies will be needed If there should be an outbreak before these are validated, there will be two choices - use them without validation or follow line A or B.